hplc method development - agilent · hplc method development ... experimental design systematic...

HPLC Method Development Systematic Approach vs Random Walk

Improving the Efficiency of Method Development and

Optimization

William Champion

Agilent Technologies, Inc.

1-800-227-9770, opt 3, opt 3, opt 2

[email protected]

Oct 3, 2012

Improving HPLC Separations

Agilent Restricted

OBJECTIVE

• Demonstrate a systematic approach to method development

• Improve understanding of separation process

• Development of more robust methods

• More efficient than “random walk”

Oct 3, 2012


Agilent Restricted

OUTLINE

• Definitions - Column Parameters

• Systematic Approach

- Comments

- Experimental Designs

- Example

Oct 3, 2012


Agilent Restricted

Some Basic Equations of

Chromatography

• Retention Factor (k)

• Selectivity or Separation Factor (α)

• Column Efficiency as Theoretical Plates (N)

• Resolution (Rs)

Oct 3, 2012


Agilent Restricted

Definition of Resolution

Rs = tR-2 - tR-1

(w2 + w1)/2

= ∆tR

w

Resolution is a measure of the ability to separate

two components

Oct 3, 2012


Agilent Restricted


Agilent Restricted

Oct 3, 2012

N = Column Efficiency – Column length and particle size

a = Selectivity – Mobile phase and stationary phase

k = Retention Factor – Mobile phase strength

Resolution … Determined by 3 Key Parameters –

Efficiency, Selectivity and Retention

The Fundamental Resolution Equation

w

∆tR =

k

N

a


Agilent Restricted

Oct 3, 2012

k

α

N

Factors that Improve Resolution

Increase

retention

Change relative

peak position

Reduce peak

width

…

.

.

w

∆tR =


Agilent Restricted

Chromatographic Profile

Equations Describing Factors Controlling RS

k = (tR-t0)

t0

α = k2/k1

N = 16(tR / tW)2

Theoretical Plates-Efficiency

Selectivity

Retention Factor

Oct 3, 2012


Agilent Restricted



k = (tR-t0)

t0

α = k2/k1

N = 16(tR / tW)2


Selectivity

Retention Factor

Oct 3, 2012

= tS

tM


Agilent Restricted



k = (tR-t0)

t0

α = k2/k1

N = 16(tR / tW)2


Selectivity

Retention Factor

Oct 3, 2012

= tS

tM

k = 1 to 20 - OK; k = 3 to 10 - Better; k = 5 to 7 - Ideal


Agilent Restricted



k = (tR-t0)

t0

α = k2/k1

N = 16(tR / tW)2

tW = tR x (4/√N)


Selectivity

Retention Factor

Oct 3, 2012

OUTLINE

• Definitions - column parameters


- Comments


- Example

Oct 3, 2012


Agilent Restricted

Experiments Are “Front-end Loaded”

Experiments Provide Easier to Interpret

Results

Insight into how to adjust conditions to

affect separation

Pre-validation

SYSTEMATIC APROACH

Oct 3, 2012


Agilent Restricted

“Front-end Loaded” Experiments:

- Know what experiments to perform

- Often can be run in groups or block (e.g.

overnight)

- Can be intimidating because it looks

like a lot of work

- More complicated designs looks like

statistics

SYSTEMATIC APROACH

Oct 3, 2012


Agilent Restricted

Results Easier to Interpret:

- Indicate where optimum is/is not

- Provides insight into trade-offs

- Tells about Sensitivity/Robustness to small

parameter changes

- What conditions are important/what are not

important (e.g., buffer)

- Easier to follow confusing interactions

(“connect the dots”)

SYSTEMATIC APROACH

Oct 3, 2012


Agilent Restricted

- Uncoordinated series of experiments

- Poor understanding of the interaction of

parameters

- Can achieve acceptable separation - but not

understand “why”

- Do not know if complicated conditions are

necessary

- Does not provide insight into sensitivity of

modifications to the conditions

(i.e., robustness)

“RANDOM WALK”

Oct 3, 2012


Agilent Restricted

Algorithm driven or search techniques

Models based upon fundamental

understanding of the chromatographic

process

Experimental design

SYSTEMATIC APROACHES

Oct 3, 2012


Agilent Restricted

Gradient screening

Isocratic separation – vary mobile phase

strength

Simple screening design

Factorial design

Experimental Designs

Oct 3, 2012


Agilent Restricted

OUTLINE



- Comments


- Example

Oct 3, 2012


Agilent Restricted

- Well understood

- Efficient, information rich

- Provides insight into fundamental behavior

- Experiments can be run in blocks

- Data easy to present and interpret

- Easy to compare effects of individual

components

- Possible to determine if there is interaction

EXPERIMENTAL DESIGN

Oct 3, 2012


Agilent Restricted

Main Effects – “One Factor at a Time”

X1

X2

Oct 3, 2012


Agilent Restricted

Main Effects

%ACN

Temp 45ºC

25ºC

40% 20%

Oct 3, 2012


Agilent Restricted

Main Effects

%ACN

Temp 45ºC

25ºC

40% 20%

Number of Exp’s = 2 x n = 4

Oct 3, 2012


Agilent Restricted

Main Effects

%ACN

Temp 45ºC

25ºC

40% 20%

Number of Exp’s = 2 x n = 6

%Acid 0%

1%

Oct 3, 2012


Agilent Restricted

- Few experiments

- Adding a factor adds two experiments

- Isolate the effect of each factor

- Data easy to present and interpret

- Good screening design

- Unfortunately, does not provide information

on interaction of factors

Single Factor Design

Oct 3, 2012


Agilent Restricted

Factorial Design for Two Factors

%ACN

Temp

30ºC

40ºC

25% 35%

Oct 3, 2012


Agilent Restricted


%ACN

Temp

30ºC

40ºC

25% 35%

Provides information at Different Levels

Oct 3, 2012


Agilent Restricted


%ACN

Temp

30ºC

40ºC

25% 35%

Looking for Interaction of Factors

Effect of Temperature at 25% and 35% ACN

Oct 3, 2012


Agilent Restricted


%ACN

Temp

30ºC

40ºC

25% 35%

… And Effect of ACN at 30º and 40ºC

Looking for Interaction of Factors

Oct 3, 2012


Agilent Restricted

Factorial Design

%ACN

Temp

30ºC

40ºC

25% 35%

For More Detailed Understanding

30%

35ºC

Oct 3, 2012


Agilent Restricted


%ACN

Temp

30ºC

40ºC

25% 35%

With Center-Point - Orthogonal Design

30%

35ºC

Oct 3, 2012


Agilent Restricted

Non-orthogonal Design for Two Factors

%ACN

Temp

30ºC

40ºC

25% 35%

(25%, 30ºC), (30%, 35ºC) and (35%, 40ºC)

30%

35ºC

Oct 3, 2012


Agilent Restricted

Factorial Design for Three Factors

with Center Point

% Acid % ACN

Temp

3 Factors, 23 + 1 = 9

Oct 3, 2012


Agilent Restricted

OUTLINE



- Comments


- Example

Oct 3, 2012


Agilent Restricted

Decide upon the objective

What are the properties of the analyte(s)

Decide upon detector, column, mp, etc.

Literature search?

Perhaps perform a few familiarization

experiments to get feel for behavior?

Gradient (e.g., 10% to 90% ACN/20 min)

Plot effect of mobile phase on retention

Perform more detailed exp’s if necessary

PROCEDURE

Oct 3, 2012


Agilent Restricted

- What is the objective?

(assay, det. impurities, prep, trace level)

- What is important?

(time, peak capacity, LOQ/LOD)

- What kinds of constraints?

(equipment, sample properties)

Decide Upon the Objective

Oct 3, 2012


Agilent Restricted

The Analytes - Benzodiazepines

Nitrazepam

Diazepam

Oxazepam Temazepam Alprazolam

Flunitrazepam Clonazepam Lorazepam

Oct 3, 2012


Agilent Restricted

http://www.sigmaaldrich.com/thumb/structureimages/03/mfcd00057903.gif



- Column (e.g., C18, phenyl, CN, etc.)

- Mobile Phase Solvents (e.g., MeOH, ACN)

- Acid (e.g., HOAc, Formic, TFA, H3PO4, none)

- Buffer (e.g., acetate, formate, none)

- Temperature

Decide upon Detector, Column, MP, etc.

Oct 3, 2012


Agilent Restricted

Model Compounds

Single Column

(Poroshell 120 EC-C18, 3.0 x 100 mm)

Single Organic Solvent

(prefer ACN)

Simple Acid/Buffer

(formic acid, ammonium formate, H3PO4)

Separation of Eight Benzodiazepines

Oct 3, 2012


Agilent Restricted


What are the analyte(s) properties

Decide upon detector, column, mobile phase, etc.

Literature search?

Perhaps perform a few familiarization experiments

to get feel for behavior?



Perform more detailed experiments if necessary

PROCEDURE

Oct 3, 2012


Agilent Restricted

Gradient screening


strength


Factorial design


Oct 3, 2012


Agilent Restricted

Initial Gradient 10% to 90% ACN/0.1% Formic Acid in 20 min

Oct 3, 2012


Agilent Restricted

Initial Gradient 10% to 90% ACN/0.1% FA in 20 min

Oct 3, 2012


Agilent Restricted

Gradient – Compare %Formic Acid 10% ACN to 90% in 20 min

No FA

0.1% FA

1% FA

Oct 3, 2012


Agilent Restricted


No FA

0.1% FA

1% FA

60%

Oct 3, 2012


Agilent Restricted


What are the properties of the analyte(s)?

Decide upon detector, column, mobile phase, etc.

Literature search?






PROCEDURE

Oct 3, 2012


Agilent Restricted

Gradient screening


strength


Factorial design


Oct 3, 2012


Agilent Restricted

Isocratic Runs – 35°C, no acid

45%

40%

35%

30%

25%

Oct 3, 2012


Agilent Restricted


45%

40%

35%

30%

25%

0

5

10

15

20

25

30

20 25 30 35 40 45 50

tR

% ACN

Retention of Peak-7

tR-7

Oct 3, 2012


Agilent Restricted


45%

40%

35%

30%

25%

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0

5

10

15

20

25

30

20 25 30 35 40 45 50

tR

% ACN

Retention of Peak-7

tR-7

ln(k-7)

ln(k) = ln(kw) – S(%B)

Oct 3, 2012


Agilent Restricted


45%

40%

35%

30%

25% Wrap-around peak

Oct 3, 2012


Agilent Restricted

Isocratic Runs – 25% ACN, 35°C, no acid

1st Inj

Wrap-around peak

2nd Inj

Gradient

Oct 3, 2012


Agilent Restricted


What are the properties of the analyte(s)

Decide upon detector, col’m, mobile phase, etc.

Literature search?






PROCEDURE

Oct 3, 2012


Agilent Restricted

Gradient screening


strength


Factorial design


Oct 3, 2012


Agilent Restricted

Screening for Main Effects/Star Points

%ACN

Temp 45ºC

25ºC

40% 20%

Oct 3, 2012


Agilent Restricted

Screening for Main Effects (Star Points)

Exp % ACN % FA Temp mM AF

1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 10

Oct 3, 2012


Agilent Restricted

Screening for Main Effects


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 10

Oct 3, 2012


Agilent Restricted


250 mL ACN 1% FA 0.1 M AF HOH

Exp mL mL mL mL

Cp 75 25 25 125

1% FA 75 2.5 25 147.5

0% FA 75 0 25 150

25 mM 75 25 62.5 87.5

10 mM 75 25 25 125

* mL Formic Acid

*

Oct 3, 2012


Agilent Restricted


1 L ACN 1% FA 0.1 M AF HOH

Exp mL mL mL mL

Cp 300 100 100 500

1% FA 300 10.0 100 590

0% FA 300 0 100 600

25 mM 300 25 250 425

10 mM 300 25 100 575

*

* mL Formic Acid

Oct 3, 2012


Agilent Restricted



1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 10

Oct 3, 2012


Agilent Restricted


No FA

0.1% FA

1% FA

Oct 3, 2012


Agilent Restricted

Compare %FA

1% FA

No FA

0.1% FA

30% ACN, 10 mM AF, 35°C

Oct 3, 2012


Agilent Restricted

Exp-1

Exp-4

Exp-5

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Oct 3, 2012


Agilent Restricted

Effect of Temp

40°C

30°C

35°C

30% ACN/0.1% FA, 10 mM AF

Oct 3, 2012


Agilent Restricted

Exp-1

Temperature – Main Effect

35% ACN, 0.1% FA 40°C

15°C

20°C

25°C

30°C

35°C

Oct 3, 2012


Agilent Restricted

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 10

Oct 3, 2012


Agilent Restricted

Effect of Ammonium Formate

30% ACN, 0.1% FA 35°C 25 mM

No AF

Oct 3, 2012


Agilent Restricted

Exp-8

Exp-9

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Oct 3, 2012


Agilent Restricted

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Yes

Oct 3, 2012


Agilent Restricted

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Yes Yes

Oct 3, 2012


Agilent Restricted

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Yes Yes Yes

Oct 3, 2012


Agilent Restricted

Star Points


1 30 0.1 35 10

2 40 0.1 35 10

3 20 0.1 35 10

4 30 1.0 35 10

5 30 0 35 10

6 30 0.1 45 10

7 30 0.1 25 10

8 30 0.1 35 25

9 30 0.1 35 0

Yes Yes Yes Defer

Oct 3, 2012


Agilent Restricted

Screening - Star Points

Conclusions

• %ACN, Temp, FA, all affect the separation

• Order of elution reversals are observed -

potentially very confusing

• Will choose to leave AF out of optimization

• FA affects separation, but not required for

chromatography

• Sometimes find our conditions or very close

using screen, but not in this case

Oct 3, 2012


Agilent Restricted

Gradient screening


strength


Factorial design


Oct 3, 2012


Agilent Restricted

Time/Effort

• Initial Gradients: > Half Day

• Isocratic: > Half Day (overnight?)

• Screening: Set-up ~ 2 hrs

• Screening: Run overnight

Oct 3, 2012


Agilent Restricted

Factorial Design for Three Factors

with Center Point

% Acid % ACN

Temp

3 Factors, 23 + 1 = 9

Oct 3, 2012


Agilent Restricted

Factorial Design

Exp % ACN % FA Temp

1 30 0.10 35

2 35 0.15 40

3 35 0.15 30

4 35 0.05 40

5 35 0.05 30

6 25 0.15 40

7 25 0.15 30

8 25 0.05 40

9 25 0.05 30

Oct 3, 2012


Agilent Restricted

Displaying Results

Oct 3, 2012


Agilent Restricted

Displaying Results

%ACN

% Acid

0.05%

0.15%

25% 35%

__ (+) 40°C

__ (-) 30°C

__ (+) 40°C

__ (-) 30°C

Oct 3, 2012


Agilent Restricted

40°C

0.15% Formic Acid, 35% ACN

30°C

Oct 3, 2012


Agilent Restricted

Exp-2

Exp-3

40°C

0.05% Formic Acid, 35% ACN

30°C

Oct 3, 2012


Agilent Restricted

Exp-4

Exp-5

40°C, 0.15%

Temp, Formic Acid, 35% ACN

30°C, 0.15%

30°C, 0.05%

40°C, 0.05%

Oct 3, 2012


Agilent Restricted

Displaying Results

%ACN

% Acid

0.05%

0.15%

25% 35%

40°C (+) __

30°C (-) __

40°C (+) __

30°C (-) __

Oct 3, 2012


Agilent Restricted

0.15% Formic Acid; 25% ACN

40°C, 0.15%

30°C, 0.15%

<------- Merged peaks

Oct 3, 2012


Agilent Restricted

Exp-6

Exp-7


40°C, 0.15%

30°C, 0.15%

Merged peaks

Oct 3, 2012


Agilent Restricted

Exp-6

Exp-7

Resolved

peaks


40°C, 0.05%

30°C, 0.05%

Oct 3, 2012


Agilent Restricted

Exp-8

Exp-9


40°C, 0.05%

30°C, 0.05%

Merged peaks

Oct 3, 2012


Agilent Restricted

Exp-8

Exp-9

Partial Sep’n

40°C, 0.15%


30°C, 0.15%

30°C, 0.05%

40°C, 0.05%

Oct 3, 2012


Agilent Restricted


40°C, 0.15%

30°C, 0.15%

35°C, 0.15%

Oct 3, 2012


Agilent Restricted


50°C, 0.15%

35°C, 0.15%

Oct 3, 2012


Agilent Restricted

40°C, 0.15%


35°C

30°C

Oct 3, 2012


Agilent Restricted

25°C

OBSERVATIONS

• Effect of small changes in acid

• Order of elution changes with % ACN

• Order of elution changes with Temp

• Similar polarity – similar retention

• 22% ACN, 30°C and 25°C, 0.15% formic acid provide minimal separation

• Expect changes in retention

Oct 3, 2012


Agilent Restricted

Gradient screening


strength


Factorial design

Experimental Designs –

Increasing “Power”

Oct 3, 2012


Agilent Restricted

Time/Effort

• Initial Gradients: > Half Day

• Isocratic: > Half Day (overnight?)

• Screening: Set-up ~ 2 hrs


• Factorial: Set-up ~ 2 hrs


Oct 3, 2012


Agilent Restricted

OBJECTIVE

• Demonstrate a systematic approach to method development

• Improve understanding of separation process

• Development of more robust methods

• More efficient than “random walk”

Oct 3, 2012


Agilent Restricted

Thank you – Questions?


Agilent Restricted

Oct 3, 2012

Bill Champion

800-227-9770, opt 3, LC Column Support

[email protected]

hplc method development - agilent · hplc method development ... experimental design systematic...

Documents